Beginning in January 2013, the American Community Survey (ACS) introduced an Internet response option. The ACS added this option alongside its traditional data collection operations: mail, computer-assisted telephone interviewing (CATI) and computer-assisted personal interviewing (CAPI). This evaluation focuses on two issues related to the addition of the Internet operation.

The first issue involves multiple responses in the ACS. There has always been the possibility that a sample address may respond more than once; for example, a household may complete a telephone interview while also returning a mail questionnaire. The Internet and mail modes are the focus of the first month of data collection, with CATI in the second month, and CAPI in the third month. However, the ACS still accepts Internet and mail responses throughout the three-month data collection period. With the addition of the Internet option, there are now more ways sampled households can submit multiple responses. The ACS needs to be aware of how often we receive multiple responses and if the addition of the Internet option increased the incidence of multiple responses because they are indications of burden on respondents and may be the result of respondents'€™ confusion. The second issue involves vacant housing units. The mail questionnaire does not give respondents the option to report the unit as vacant. Therefore, the ACS always followed up with mail returns for potentially vacant units to confirm the status of the unit. However, the Internet instrument asks the same questions regarding household members and residency as the CATI and CAPI instruments. This means that Internet respondents may identify the sample address as a vacant unit. The ACS currently requires follow-up with vacant units identified via the Internet in order to verify their unit status. This report provides the information ACS managers need to consider any change(s) related to follow-up data collection for vacant units identified in the Internet instrument.